All states have in place laws that require persons to report suspected child maltreatment to public child protection agencies, and over two million reports are made in the United States every year. However, there are strong social, cultural economic and systemic forces operating to suppress the reporting of suspected maltreatment, and some authorities argue that as few as 20% of suspected cases are reported. It is postulated that a child's situation in his or her family must be perceived to be intolerable in order to overcome the community's reluctance to make a maltreatment report. If such is the case, a maltreatment report may be an indicator for a general condition of risk to a child, even though the reported situation may fail to meet the legal definition of child abuse or neglect. Children who are reported for maltreatment have been identified by and are accessible to a primary community human services delivery agency. Should it prove that children reported for maltreatment early in life are at increased risk of dying from all causes, then reporting could provide both an early indicator of severe risk and an access point for infant and child mortality prevention and risk reduction. The specific aim of this study is to determine if children, especially minority children, reported for child maltreatment in the first five years of life, are at greater risk of dying during that period than children who are not reported. A birth cohort consisting of all N.C. children for whom there are birth certificates on file for 1985 and 1986 will be stratified on the basis of race and having been reported or not reported for child maltreatment during the first five years of life. The cohort will then be followed, retrospectively, through the fifth year of life in order to determine if reported children, especially minority children are at greater risk of dying than children who were not reported.